El‐Khayat 2015.

Study characteristics
Methods	Methods: randomised controlled trial, 2 groups, set in Kasr Al‐Aini Teaching Hospital at Cairo University and a Middle East IVF Centre, Egypt February 2012 to October 2014 Number of participants randomised: 332 Number of participants analysed: 332
Participants	Inclusion criteria: women with unexplained infertility or couples with mild male factor infertility; female partner younger than 39 years; regular menstrual cycles; body mass index < 32 kg/m²; normal uterine cavity with normal thin endometrium measuring < 5 mm on Day 4; bilateral tubal patency (demonstrated by laparoscopy or hysterosalpingography); normal hormonal profile Exclusion criteria: women diagnosed with infertility due to other causes; significant cardiovascular, pulmonary, renal, neurological, or hepatic problems; presence of ovarian cyst > 2 cm before stimulation; abnormal endometrial cavity due to submucous myoma; endometrial polyp; intrauterine synechia; septate or bicornate uterus Cause of infertility: unexplained infertility, mild male factor
Interventions	Intervention group: endometrial scratching and office hysteroscopy between Days 4 and 7 of the menstrual cycle with the vaginoscopic ‘no touch technique’ Control group: office hysteroscopy between Days 4 and 7 of the menstrual cycle with the vaginoscopic ‘no touch technique’ Both groups: ovarian stimulation consisted of clomiphene citrate 100 mg/d from Day 3 to 7, human menopausal gonadotropin 75 IU/d from Day 6 to 8. Transvaginal ultrasound was done on Day 9, and when 2 to 3 follicles with > 18 mm diameter were present, human chorionic gonadotropin trigger of 10,000 IU was administered. Intrauterine insemination (IUI) was performed 36 hours after the trigger Degree of endometrial injury: grasping forceps with teeth Timing of endometrial injury: follicular phase (Day 4 to 7) of the preceding cycle Study length: 1 cycle Type of conception: IUI
Outcomes	Reported in the paper: Live birth rate (not defined) Clinical pregnancy rate defined as presence of intrauterine gestation with foetal heart pulsations demonstrated by transvaginal ultrasound at 6 to 7 weeks' duration Abortion (miscarriage) rate (not defined) Multiple pregnancy rate Presence or absence of significant pain recorded, but this does not fit the criteria for the outcome 'pain' in this review
Notes	Funding source: none Conflicts of interest: "all authors have nothing to disclose" Trial registration: NCT01544426 (retrospectively registered) Author correspondence was undertaken
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "computer generated random number tables"
Allocation concealment (selection bias)	Low risk	Quote: "opaque sealed envelopes containing the participants' group allocation" The random allocation was put into envelopes every "24 hours at a location different from the study site and sent to an assigned nurse who opened each envelope just before the office hysteroscopy" Study authors confirmed via correspondence that envelopes were sequentially numbered and revealed that this was a mechanism to help ensure no violation of allocation concealment
Blinding of participants (performance bias)	Low risk	The paper stated, "the patients were blinded to group allocation" Participants were undergoing either hysteroscopy or hysteroscopy and endometrial injury. Although no anaesthesia or analgesia was used, and participant blinding was not formally tested, the control procedure is likely to simulate the intervention and therefore is likely to have blinded participants to their allocation
Blinding of personnel (performance bias)	High risk	Study did not report any blinding of personnel
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Study did not report blinding of outcome assessors, and it was unlikely; however outcomes were unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Low risk	Four participants from the intervention group and 2 from the control group were lost to follow‐up, and none discontinued interventions. The study reported the number of participants missing, and it was similar between groups. The study authors performed intention‐to‐treat analysis
Selective reporting (reporting bias)	Unclear risk	Trial was registered retrospectively. We confirmed with trial authors that pain was not recorded
Other bias	Low risk	We did not identify any other potential sources of bias